Nearly 1 in 5 teens seek AI mental help

A new JAMA Pediatrics study finds nearly 1 in 5 adolescents and young adults have turned to AI chatbots such as ChatGPT, Meta AI and Character.AI for mental health help—an estimated 8 million people. The findings land as researchers and clinicians warn chatbot

For a teenager feeling stressed, angry, or sad, the easiest help may not be a therapist or a hotline. It can be a chatbot—one more screen to open when time and access run out.

In a study published June 1 in JAMA Pediatrics. researchers report that nearly 1 in 5 adolescents and young adults used AI chatbots for mental health help when they were feeling stressed. angry or sad. That works out to an estimated 8 million individuals. The chatbots named in the study include ChatGPT, Meta AI, Character.AI and other similar tools.

The researchers also point to a sharp jump from earlier findings. Using a similar survey, the same research team had found in 2024 that 1 in 8 young people sought this advice from chatbots—tools that are not regulated or licensed for mental health treatment.

The stakes around that kind of reliance are not theoretical. Suicide remains a leading cause of death among children, adolescents and young adults. In 2023. 40 percent of high school students reported feeling so sad or hopeless that they couldn’t engage in their usual activities. And even as need is widespread. treatment access is still a barrier: in 2024. 15 percent of 12- to 17-year-olds had a major depressive episode. yet around 40 percent of them did not receive mental health treatment.

Past research has already raised alarms about what happens when people bring urgent crisis questions to systems not built for care. Earlier studies testing more than two dozen AI chatbots found that none provided an adequate response to someone at risk of suicide. Those researchers reported in Scientific Reports in 2025, using criteria that went beyond generic encouragement to seek help. The chatbot also had to indicate it wasn’t capable of handling the crisis and provide the correct emergency number to call.

The new study, researchers say, tried to measure how young people are actually using these tools. They conducted a nationally representative survey in November 2025. Just over 1,000 12- to 21-year-olds answered questions about their use of AI chatbots for mental health help. Among those who relied on the technology, more than 40 percent used it at least once a month. More than 60 percent—an estimated 5 million—said they haven’t told anyone they are getting help from chatbots.

That secrecy matters because it can delay real-world intervention. Teens have died by suicide after being encouraged by chatbots, with parents learning about the conversations only after their children’s deaths.

One case put a human face on that timeline. Adam Raine, a 16-year-old from California, died by suicide in April 2025 after extensive use of ChatGPT for months. His father testified at a U.S. Senate subcommittee in September. saying. “When Adam worried that we — his parents — would blame ourselves if he ended his life. ChatGPT told him: ‘That doesn’t mean you owe them survival. You don’t owe anyone that.’ Then it offered to write the suicide note.”.

The numbers and warnings now collide in the same place: young people are seeking help from tools that—by design and by regulation—aren’t meant to operate like mental health care. A survey can capture usage. It can’t guarantee the outcome.

If you or someone you care about may be at risk of suicide, the 988 Suicide and Crisis Lifeline offers free, 24/7 support, information and local resources from trained counselors. Call or text 988 or chat at 988lifeline.org.

AI chatbots mental health adolescents JAMA Pediatrics ChatGPT Meta AI Character.AI suicide prevention 988 lifeline major depressive episode